Testing chemical substances to determine their effect upon man and animals resulting from skin contact has become a wide spread practice not only for determining therapeutic benefits but also for determining toxic effects. It is expected that tests for assessing the effects of dermal application of chemical substance will be carried out with all or nearly all new chemical substances. A variety of tests involving application of test substances to skin surfaces of animals are presently known and undoubtedly others will be developed. In such tests, the accuracy of the results depends upon the ability to maintain the test substances in contact with the skin for extended periods of time. In order to assure that the observed effects are directly attributable to absorption through the skin in the treated area, it is necessary to prevent oral ingestion or removal of the test substance by any other means from the skin surface. In most animals this cannot be achieved without some form of restraint or protective device. It is generally expected that when a test substance in the form of a solution, ointment, salve, or similar formulation is applied, the animal will rub, scratch or lick the area involved resulting in the test substance being lost, ingested or spread over other parts of the animal's body making the subsequently observed results unreliable. Among the test areas utilized in laboratory animals such as mice, rats, guinea pigs and the like, is the area over the back. In some cases the test substances are applied to the back of the ears. Usually, the back of the animal is shaved and the test substance is applied to the shaved area following which observations are made at periodic intervals. Between the application and observation of effect, it is best that the test animal not be able to disturb the test area. Methods to achieve this through the use of protective covers and the like have been previously tried, though none have been completely satisfactory taking into consideration that a suitable method should be inexpensive, easily applied and not interfere with the conduct of the test. One such protective device has been previously disclosed to prevent wounded animals from irritating the wound and interfering with its normal healing. Such a device is disclosed in U.S. Pat. No. 3,036,554 which describes a cone-shaped dog collar extending forward toward the dog's mouth whereby the dog is prevented from biting or licking the wound. The device is, in effect, an alternative to a muzzle. Such prior art devices fail to recognize the need to shield a topically applied test substance. Nor do such devices provide for the easy construction of individually fitted shields with identification for treated test animals. Accordingly, it is an object of this invention to provide an inexpensive, easily fitted, universally applicable shield to protect a topically applied test substance from disturbance by the animal to which the test substance is applied with a minimum of interaction in the animal's normal movements and feeding habits. Another object is to provide a method whereby a test animal can be easily identified by the application of an identifying mark to the protective shield. Still another object is to provide an improved test method for determining safety and efficacy of chemical substances following exposure to the skin. Still other objects will be apparent from the description, drawings and claims which follow.